. The long-tern objective of this research is to determine the extent to which human genetic variability contributes to TB susceptibility and to the character of the resulting disease. To avoid pitfalls associated with previous work in this field, these questions will be examined at the nucleotide level in approximately 250-300 pediatric patients assigned to precisely defined clinical disease categories. The study population will be pediatric TB patients living in Houston, TX, a city with the highest reported rate of childhood TB in the U.S. Genotyping strategies, already employed successfully in the laboratories of this international collaborative study group will be used to determine the association of TB in pediatric patients with polymorphisms in four candidate susceptibility loci: NRAMPI (encoding a macrophage membrane protein critical in mediating murine susceptibility to BCG infection); TNFA (encoding the cytokine tumor necrosis factor-alpha); MBL (encoding mannose binding protein); and VDR (encoding the Vitamin D receptor). New, state-of-the-art molecular strategies will also be used to conduct a genome-wide screen to identify loci not previously associated with human TB susceptibility. All M.tuberculosis case isolates will be analyzed molecularly to test the hypothesis that distinct pathogen genotypes are non-randomly associated with certain host genotypes (i.e., the potential role of host-pathogen genotypic combinations in mediating disease character will be examined). By wedding analysis of host and pathogen molecular data with clinicopathologic correlates of afflicted patients, the proposed study is intended to contribute new insights into host-pathogen relationships.